Add image converters for WMF using LibreOffice and ImageMagick

2 files changed, 80 insertions, 5 deletions

diff --git a/mammoth/documents.py b/mammoth/documents.py
index b1e2c1e..a6e125c 100644
--- a/mammoth/documents.py
+++ b/mammoth/documents.py
@@ -69,12 +69,11 @@ class Tab(Element):
     pass
     
 
-@cobble.visitable
+@cobble.data
 class Image(Element):
-    def __init__(self, alt_text, content_type, open):
-        self.alt_text = alt_text
-        self.content_type = content_type
-        self.open = open
+    alt_text = cobble.field()
+    content_type = cobble.field()
+    open = cobble.field()
 
 
 def document(children, notes=None, comments=None):
diff --git a/mammoth/images.py b/mammoth/images.py
index 9dba353..9cc358c 100644
--- a/mammoth/images.py
+++ b/mammoth/images.py
@@ -1,4 +1,9 @@
 import base64
+import io
+import os
+import shutil
+import subprocess
+import tempfile
 
 from . import html
 
@@ -25,3 +30,74 @@ def data_uri(image):
     return {
         "src": "data:{0};base64,{1}".format(image.content_type, encoded_src)
     }
+
+
+_wmf_extensions = {
+    "image/x-wmf": ".wmf",
+    "image/x-emf": ".emf",
+}
+
+
+def libreoffice_wmf_conversion(post_process=None):
+    if post_process is None:
+        post_process = lambda x: x
+    
+    def convert_image(image):
+        wmf_extension = _wmf_extensions.get(image.content_type)
+        if wmf_extension is None:
+            return image
+        else:
+            temporary_directory = tempfile.mkdtemp()
+            try:
+                input_path = os.path.join(temporary_directory, "image" + wmf_extension)
+                with io.open(input_path, "wb") as input_fileobj:
+                    with image.open() as image_fileobj:
+                        shutil.copyfileobj(image_fileobj, input_fileobj)
+                
+                output_path = os.path.join(temporary_directory, "image.png")
+                subprocess.check_call([
+                    "libreoffice",
+                    "--headless",
+                    "--convert-to",
+                    "png",
+                    input_path,
+                    "--outdir",
+                    temporary_directory,
+                ])
+                
+                with io.open(output_path, "rb") as output_fileobj:
+                    output = output_fileobj.read()
+                
+                def open_image():
+                    return io.BytesIO(output)
+                
+                return post_process(image.copy(
+                    content_type="image/png",
+                    open=open_image,
+                ))
+            finally:
+                shutil.rmtree(temporary_directory)
+    
+    return convert_image
+
+
+def imagemagick_trim(image):
+    command = ["convert", "-", "-trim", "-"]
+    process = subprocess.Popen(command, stdin=subprocess.PIPE, stdout=subprocess.PIPE)
+    try:
+        with image.open() as image_fileobj:
+            shutil.copyfileobj(image_fileobj, process.stdin)
+        output, err_output = process.communicate()
+    except:
+        process.kill()
+        process.wait()
+        raise
+        
+    return_code = process.poll()
+    if return_code:
+        raise subprocess.CalledProcessError(return_code, command)
+    else:
+        def open_image():
+            return io.BytesIO(output)
+        
+        return image.copy(open=open_image)